An Artificial Immune Clustering Approach to Unsupervised Network Intrusion Detection

To solve the problem of existing artificial immune network-based intrusion detection model, an unsupervised network intrusion detection method based on Adaptive Radius Immune Algorithm (ARIA) is presented in this paper. ARIA and graph clustering algorithm are employed to generate detectors. The obtained results suggest that this method achieves higher detection rate and lower false positive rate over KDD Cup 1999 data set, and is more effective than other intelligent clustering and classification approaches such as artificial immune network-based and SVM-based intrusion detection models.